PERSONALITY PROCESSES AND INDIVIDUAL DIFFERENCES
Effects of Optimism, Pessimism, and Trait Anxiety on Ambulatory Blood
Pressure and Mood During Everyday Life
University of Helsinki
Karen A. Matthews, Janine D. Flory,
Jane F. Owens, and Brooks B. Gump
University of Pittsburgh
This study tested whether disposition^ measures of optimism, pessimism, and anxiety affected ambu-
latory blood pressure (BP) and mood and whether any cardiovascular effects of dispositions were
moderated by mood. Pessimistic and anxious adults had higher BP levels and felt more negative and less
positive than did optimists or low anxious adults throughout the monitoring. The few times that optimists
did feel negative were associated with levels of BP as high as those observed among pessimists or
anxious individuals, regardless of their mood. To the extent that trait anxiety measures neuroticism, these
findings suggest that neuroticism is directly related to health indicators rather than simply to illness
behavior. Furthermore, the results suggest that pessimism has broad physiological and psychological
Optimists are people who expect positive outcomes. As a con-
sequence, they expect to cope effectively with everyday stress and
challenge, whereas pessimists are those who expect negative out-
comes and do not expect to cope successfully (for reviews, see
Scheier & Carver, 1985, 1987, 1992). Optimists are likely to
persist in their goal-directed efforts, whereas pessimists are more
likely to withdraw effort, become passive, and potentially give up
on achieving their goals. As such, pessimists are hypothesized to
be more likely to experience the physical and emotional conse-
quences of stressful situations than are optimists.
In the past, optimists and pessimists have been classified by high
versus low scores on a single dimension and treated as bipolar
Katri Raikkonen, Department of Psychology, University of Helsinki,
Helsinki, Finland; Karen A. Matthews, Janine D. Flory, Jane F. Owens, and
Brooks B. Gump, Department of Psychiatry, University of Pittsburgh.
Katri Raikkonen is now on sabbatical from the University of Helsinki
and is supported by the Finnish Academy as a visiting scientist at the
Department of Psychiatry, University of Pittsburgh. Brooks B. Gump is
now at the Department of Psychology, State University of New York
College at Oswego.
This work was supported by National Institutes of Health Grants
HL38712 and HL07560, a visiting scientist award (4469) from the Finnish
Academy, and the Ella and Georg Ehrnrooth Foundation.
We thank Michael F. Scheier for his comments on an earlier version of
this article. The statistical expertise of Pertti Keskivaara is gratefully
Correspondence concerning this article should be addressed to Karen A.
Matthews, Department of Psychiatry, University of Pittsburgh, 3811
O'Hara Street, Pittsburgh, Pennsylvania 15213. Electronic mail may be
sent to firstname.lastname@example.org.
opposites. Indeed, in most contexts in which they have been
studied, the effects of optimism and pessimism have been com-
plementary (Carver et al., 1993). However, more recent research
suggests that they are not always two sides of the same coin; for
example, they can have different correlates and tend to emerge as
separate factors (Mroczek, Spiro, Aldwin, Ozer, & Bosse, 1993;
Robinson-Whelen, Kim, MacCallum, & Kiecolt-Glaser, 1997).
This issue is not yet resolved, so these findings have led research-
ers to suggest that the effects of optimism and pessimism be
evaluated separately as a matter of course (Scheier, Carver, &
A relative paucity of data exists bearing on whether pessimism
and optimism, either as a bipolar trait or as two independent
dimensions, lead people to experience different physical health
consequences in everyday stressful situations. After coronary by-
pass surgery, optimists have been shown to recover more quickly
from surgery and to have less severe anginal pain than pessimists
(Fitzgerald, Tennen, Affleck, & Pransky, 1993; Scheier et al.,
1989). Also, optimists have been shown to report fewer physical
health complaints than pessimists (Robbins, Spence, & Clark,
1991; Scheier & Carver, 1985; Scheier et al., 1994). Conversely, it
has been shown that pessimists tend to display greater diastolic
blood pressure (DBP) reactivity to a laboratory stressor (Williams,
Riels, & Roper, 1990). In the only study of physical health that
examined optimism and pessimism separately, pessimistic cancer
patients were more likely to die during the follow-up than their less
pessimistic counterparts (Schulz, Bookwala, Knapp, Scheier, &
Williamson, 1996). Optimism was unrelated to survival.
Ambulatory blood pressure (BP) is an important physical health
outcome. Average ambulatory BP is more closely associated with
target-organ damage than are casual measures obtained in clinical
Journal of Personality and Social Psychology, 1999, Vol. 76, No. 1, 104-113
Copyright 1999 by the American Psychological Association, Inc. 0022-3514/99/$3.00
OPTIMISM AND AMBULATORY BLOOD PRESSURE
assessments. For example, ambulatory measurements of BP have
been shown to improve the prediction of left ventricular mass
(Devereux et al., 1983; Sokolow, Werdegar, Kain, & Hinman,
1966) and cardiovascular mortality (Perloff, Sokolow, & Cowan,
1983). In the present study, we tested the hypothesis that pessi-
mists, as defined by high scores on the Life Orientation Test (LOT;
Scheier & Carver, 1985), would experience elevated ambulatory
BP throughout the day, even if they were normotensive and in
good health. We also evaluated the association of LOT pessimism
and optimism subscales separately. In our study, ambulatory BP
was assessed approximately every 30 min during waking hours
on 2 workdays and 1 nonworkday in 100 normotensive adults. At
the time of each BP measurement, participants completed a diary
about other possible determinants of BP (e.g., physical activity,
posture, and caffeinated beverage intake; Schwartz, Warren, &
Pickering, 1994) so that the influences of optimism and pessimism
could be examined independent of other common determinants
Perhaps also relevant to understanding elevations in ambulatory
BP is another key disposition, neuroticism or negative affectivity
(Watson & Clark, 1984). This refers to a broad, stable dimension
of personality consisting of chronic negative emotions, including
sadness, anxiety, guilt, and anger. The Spielberger Trait Anxiety
Inventory (Spielberger, 1983) is considered a proxy for neuroti-
cism (Smith, Pope, Rhodewalt, & Poulton, 1989; Watson & Clark,
1984), and its scores are highly correlated (r = .73; Watson &
Clark, 1984) with scores on Eysenck and Eysenck's (1968) Neu-
roticism Scale. Trait anxiety scores predicted BP change over time
in a sample of healthy middle-aged women (Markovitz, Matthews,
Wing, Kuller, & Meilahn, 1991), and a conceptually related score,
the Framingham tension score, predicted new hypertension among
middle-aged men followed for 18-20 years (Markovitz, Matthews,
Kannel, Cobb, & D'Agostino, 1993). To our knowledge, no study
in normotensives taking into account the common determinants of
BP has evaluated the effect of trait anxiety on ambulatory BP.
Therefore, a second study objective was to evaluate whether anx-
iety, as assessed by the Spielberger Trait Anxiety Inventory, was
associated with elevated BP across the monitoring period.
The second objective has theoretical importance because of
questions about the discriminant validity of optimism (Smith et al.,
1989; cf. Marshall, Wortman, Kusulas, Hervig, & Vickers, 1992).
In two studies of healthy undergraduates, high pessimism, as
measured by the LOT, and high neuroticism, as measured by the
Spielberger Trait Anxiety Inventory, were each related to subse-
quent reports of frequent minor physical symptoms and problem-
focused coping (Smith et al., 1989). Statistical adjustments for trait
anxiety rendered nonsignificant the correlations between pessi-
mism and symptoms of problem-focused coping, whereas statisti-
cal adjustments for pessimism did not substantially alter the com-
parable anxiety correlations. This led the investigators to conclude
that the relationship between pessimism and symptoms reflect the
influence of a predisposition to experience negative affect, or a
weaker measure of the general trait. Perhaps this reasoning can
also apply to prediction of physical health outcomes, including
ambulatory BP. Mroczek et al. (1993) reported that the associa-
tions of high pessimism and low optimism with illness severity
were each statistically significant, but statistical controls for neu-
roticism reduced the optimism and pessimism associations to
nonsignificance. However, Schulz et al. (1996) reported that de-
pression was not correlated with cancer mortality, whereas high
pessimism was, in fact, correlated. Our analytic strategy permitted
examining the individual and unique effects of optimism, pessi-
mism, and anxiety on ambulatory BP.
An important determinant of ambulatory BP is mood (Schwartz
et al., 1994), and both optimism-pessimism and anxiety are cor-
related with mood states. In general, optimists are in a more
positive mood, whereas pessimists are in a more negative mood
(Marshall et al., 1992; Scheier et al., 1994). For example, pessi-
mistic women have been shown to report higher levels of depres-
sive symptoms than optimistic women (Marshall & Lang, 1990),
especially in the presence of stressful ongoing problems (Brom-
berger & Matthews, 1996) or after a stressful life change (Carver
& Gaines, 1987). Low optimism, but not high pessimism, has been
correlated with reports of daily hassles (Mroczek et al., 1993). In
general, individuals high in neuroticism are in a more negative
mood than are those low in neuroticism. For example, it has been
shown that neurotic students report increases in daily anxiety as
they approach a major examination (Bolger, 1990) and that neu-
rotic undergraduates report more daily depressive and gastrointes-
tinal symptoms across 2 months than their less neurotic counter-
parts (Larsen & Kasimatis, 1991). Some evidence suggests that
neuroticism is more highly correlated with high negative mood
than with low positive mood (Larsen & Ketelaar, 1991). Neurot-
icism is associated with high exposure to daily stressors, with
interpersonal conflicts being the most important daily stressor in
explaining the neuroticism-distress relationship (Bolger & Schill-
ing, 1991). A third objective of the present study was to extend
these findings by examining mood ratings made simultaneously
with the BP readings in light of the dispositions of the study
participants and to explore whether any obtained associations
between dispositions and ambulatory BP might be moderated by
changes in negative mood. Frequent measurement of mood states
during participants' usual daily activities provides different infor-
mation than a single measurement of current mood during extraor-
dinary or artificial circumstances or a single summary measure of
mood taken each day. We expected pessimists and neurotics to
report more negative mood and experience more frequent negative
interpersonal interactions, and we expected optimists to report
more positive mood. The analytic strategy again allowed testing
for the individual and unique effects of each personality variable.
One hundred participants (50 women and 50 men) between 30 and 45
years of age (women: M = 37.1 years, SD = 4.1; men: M = 36.5 years,
SD = 4.7) were recruited from a university community setting. Eighty-
eight percent of the participants were Caucasian, and 10% were African
American. Half of the participants were employed in professional or
managerial jobs, and the other half worked in technical or clerical posi-
tions; men and women were matched in terms of their level of occupational
prestige. All participants worked more than 20 hr per week and were
employed during daytime hours. A telephone interview screened potential
participants to ensure that they did not smoke, were free of chronic disease
(including hypertension), were less than 20% overweight, and were not
taking medications with known cardiovascular or metabolic effects (hor-
mone replacement or contraceptives). Additional eligibility criteria in-
cluded resting (laboratory) BP assessments in the normal range (systolic
RAIKKONEN, MATTHEWS, FLORY, OWENS, AND GUMP
BP [SBP]: £140 mmHg; DBP: <90 mmHg). Table 1 provides demo-
graphic and other data on the men and women in the study.
Each participant was involved in 3 days of ambulatory monitoring. The
monitor was worn on 2 working days from 8:00 a.m. to 10:00 p.m. and on 1
nonworking day from 10:00 a.m. to 6:00 p.m., with BP recorded at 30-min
intervals. Measurement on both working and nonworking days allowed
broad sampling of daily experiences at home and at work. Each time the
cuff deflated, the participants completed a page in an ambulatory diary
designed to measure concurrent behavioral and psychological states.
Before the ambulatory monitoring, all participants attended an assess-
ment session in the laboratory. At the initial visit, participants were briefed
concerning the protocol, and informed consent was obtained. At this point
of the protocol, resting BP levels, height, and weight were measured in the
laboratory, and demographic information was gathered. BP was measured
with an automated BP monitor (IBS Model SD-700A). Participants then
completed a battery of psychological questionnaires and participated in
interviews providing data on a variety of personality characteristics. After
the assessment procedure, participants were trained in the use of the
ambulatory BP monitoring device and the diary. That is, participants were
trained and provided written instructions on (a) how to take the monitor off
and (b) the correct placement of the microphone and the BP cuff. They
were also trained on how to change the battery of the monitor. A sample
BP reading was taken at this time, and participants completed a page in the
diary. Participants then left the laboratory and, within the next week (unless
their work schedule did not permit it), wore the monitor for 3 days (2
working days and on 1 nonworking day). If participants were not working
Demographic, Psychological, and Health
Characteristics of the Sample
Body weight (kg)
Body mass index (kg/m2)
Overall optimism score
Below median (£5)
Above median (>5)
Below median (=£2)
Above median (>2)
Below median (:£8.5)
Above median (>8.5)
Below median (£15)
Above median (>15)
Ambulatory systolic blood
Ambulatory diastolic blood
the next 2 consecutive days, the measurement period continued into the
following workweek. At the end of each of the 3 days, the participants were
asked to complete an "end-of-day" assessment in the diary, including their
psychological and behavioral states while wearing the monitor on that
After wearing the monitor for 3 days, participants returned to the
laboratory for a second assessment session. During this session, they
completed the remainder of the questionnaires and interviews not admin-
istered at the first session as a result of time constraints. After the second
assessment session, any questions the participant had about the study were
answered. Participants were then paid $100 for taking part in the study.
Diary of mood and events. The ambulatory diary, designed to measure
behavioral and psychological states with potential influences on cardio-
vascular activity, was adapted from the diary developed by Hedges, Krantz,
Contrada, and Rozanski (1990) for monitoring ambulatory physiological
states and mood. The participants responded to 17 words describing their
current mood state (e.g., stressed, angry, happy, or tired) on a 4-point scale
ranging from not at all (1) to a lot (4). In the present study, the averages
of the 75 diary entries for each of the 17 mood states were subjected to
principal-components factor analysis followed by varimax rotation. Three
factors with eigenvalues above 1 (4.4,2.3, and 1.3) accounted for 47.1% of
the variance. A scree plot of factor roots was consistent with the extraction
of three factors. The first factor consisted of high positive loadings for
stressed, irritable, angry, resentful, nervous, worried, impatient, and sad
and was labeled "negative mood." The second factor consisted of high
positive loadings for pleased, happy, content, energetic, in control, and
interested-involved and was labeled "positive mood." Finally, the third
factor consisted of high positive loadings for bored, apathetic, and tired and
was labeled "bored."1 The items loading on each factor were unit weighted
and summed for each of the 75 data collections to designate periods of
negative, positive, and bored mood during ambulatory monitoring. Cron-
bach alpha coefficients for the negative, positive, and bored mood scales,
as averaged across the 75 data collections, were .88, .87, and .50, respec-
tively. Total scores were used rather than single items because of the
increased reliability of measurement (cf. Hedges et al., 1990), the factor
structure underlying the 17 items, and the desire to avoid Type 1 error.
The diary also included information regarding the circumstances under
which the mood and BP assessments were made. Participants indicated
their location (home, work, or driving a car), posture (reclining, sitting, or
standing), physical activity level (low, mild, moderate, or strenuous),
substance use (alcohol, caffeine, or "over-the counter" medications), inter-
personal social interaction (yes or no), and the type of interpersonal
interaction (positive, neutral, negative, or an argument) at the time of the
The "end-of-day" assessment in the diary included information regard-
ing mood; the participants responded to the same 17 mood states that they
had rated throughout the day. However, they were instructed to respond to
these mood items in terms of how strongly they felt each mood during that
day. They were also asked to record the most stressful or difficult event that
had happened to them while wearing the monitor that day and to rate the
severity of the event on a 4-point scale ranging from very mild (1) to very
Cardiovascular measures. BP was assessed outside the laboratory
1 Separate factor analyses were run for the mood ratings of men and
women. Three factors with eigenvalues above 1 emerged on both groups,
the factors being substantively identical with the overall factor analysis.
Therefore, designation of periods of negative, positive, and bored moods
for women and men across the days of monitoring according to positive
loadings on the overall factor analysis was deemed valid.
OPTIMISM AND AMBULATORY BLOOD PRESSURE
environment with the Colin ABPM-630 ambulatory monitor. This monitor
has the major advantage of using a silent CO2 cartridge in cuff inflations
rather than an inflation pump that can produce varying degrees of noise.
Such a measurement may be less noticeable to others and, therefore, less
likely to cause participants to deviate from their normal behavior patterns
(Matthews, Owens, Allen, & Stoney, 1992). The BP cuff (child, adult, or
large adult size) was placed on the participant's nondominant upper arm
and positioned so that the microphone was over the inner aspect of the arm.
The Colin monitor allows for both auscultatory (detected via microphone)
and oscillometric determinations of BP. Oscillometric values were used in
favor of auscultatory ones because this measurement technique is less
prone to contamination if the microphone is not precisely mounted on the
artery, and thus it serves as a more reliable estimate of ambulatory BP. In
addition, the oscillometric method produced less missing data than the
Dispositional optimism. Dispositional optimism was assessed via the
LOT (Scheier & Carver, 1985). The LOT consists of eight items (along
with three filler items), of which four are worded positively (in the
optimism direction) and four are worded negatively (in the pessimism
direction). With this scale, respondents indicate to what degree they en-
dorse statements such as "I rarely count on good things happening to me"
on a scale ranging from 0 (strongly disagree) to 3 (strongly agree). After
the positively worded items had been reversed, the eight items were totaled
to yield an overall optimism score, with high scores indicating low opti-
mism (high pessimism). In addition to the overall score, the positively and
negatively worded items were totaled separately to yield optimism and
pessimism scores. The two-dimensionality of the LOT was verified by
principal-components factor analysis with varimax rotation. Indeed, the
analysis yielded two factors with eigenvalues greater than 1 (3.90
and 1.36), these factors explaining 66% of the variance. The first factor
consisted of high positive loadings of the negatively worded items, and the
second factor consisted of high positive loadings of the positively worded
items (cf. Marshall et al., 1992; Scheier et al., 1994). Cronbach alpha
coefficients for the overall score and for the optimism and pessimism
subscale scores were .84, .75, and .86, respectively (cf. Scheier & Carver,
1985; Scheier et al., 1994).
Trait anxiety. Trait anxiety was assessed with the 10 items from the
Spielberger Trait Anxiety Inventory (Spielberger, 1983). Respondents in-
dicated how they "generally" felt by endorsing statements such as "I feel
nervous and restless." Endorsements were made on a scale ranging from 1
(almost never) to 4 (almost always). The items were totaled to yield an
overall anxiety score, with high scores indicating high trait anxiety. Cron-
bach's alpha coefficient was .86 in the present sample (cf. Spielberger,
Cardiovascular measures. Approximately 16% of the readings (1,223
of 7,500) were missing because of problems with the monitor or because
the participant failed to wear the monitor during the predetermined times.
Additional readings were excluded based on criteria developed in our
laboratory for assessment of BP in normotensive, healthy populations.
Twenty-one readings were excluded because (a) DBP was greater than 140
mmHg or less than 40 mmHg or (b) pulse pressure (i.e., the arithmetic
difference between SBP and DBP) was less than 15 mmHg. Forty-two SBP
readings below 80 mmHg and 3 SBP readings above 200 mmHg were
excluded because the questionable BP was greater or less than 2 standard
deviations of the participant's average BP and inspection of physical
activity or posture (as indicated in the ambulatory diary) could not explain
an unusually high or low SBP. It should be noted that if any of these
exclusion criteria were met, all data for a particular point in time were
excluded. That is, if a DBP reading was greater than 140 mmHg, the
corresponding SBP data were also excluded from the analyses. Thus, an
average participant produced 62 (SD = 6.81) valid BP readings, the data
overall consisting of 82.8% (6,211 of 7,500) of valid observations.
Diary of mood and events. An average participant missed fewer
than 3.1% (SD = 3.2) of the mood observations and fewer than 2.6%
(SD = 2.8), 1.8% (SD = 2.4), 2.5% (SD = 3.3), 2.4% (SD = 3.2), 3.4%
(SD - 3.5), and 2.0% (SD = 2.2) of the observations on location, posture,
physical activity, substance use, interpersonal interaction, and type of
interpersonal interaction, respectively. Of the ambulatory diary observa-
tions, 85.3% (6,397 of 7,500) were available for the study. Alto-
gether, 97.4% (6,050 of 6,211) of the ambulatory BP readings had a
simultaneous daily diary observation available. There were no differences
between optimists and pessimists or between individuals high and low in
trait anxiety in the extent to which either mood or BP values were missing.
To meet the specific research questions and to benefit from the experi-
ental sampling method, we used a statistical approach combining both the
idiographic and nomothetic levels of analysis. Therefore, in addition to
simple descriptive statistics and within-subject correlations, random regres-
sion models (PROC MIXED; SAS Institute) were used to analyze the data.
Indeed, the random effects approach yields regression estimates for each
individual (on a within-subject basis) and for the group as well. The
analysis requires no equality of slopes and intercepts of all variables across
the sample and allows different error structure assumptions (including
autocorrelation effects between adjacent residuals).
Before proceeding to the study hypotheses, we determined which of the
potential covariates measured in the ambulatory diary predicted SBP and
DBP. In these analyses, repeated BP readings were used as the dependent
variables in separate models. Physical activity, posture, location, substance
use (alcohol, caffeine, "over the counter" medications), and interpersonal
social interaction (yes/no; all dummy coded), measured concurrently with
BP, served as time-varying (within-subject) covariates in each model.
Thus, our analysis determined, on a within-subject basis, the extent to
which momentary changes in BP were attributable to momentary changes
in physical activity, posture, location, substance use, and interpersonal
interaction. Each model also included gender and level of occupational
prestige (dichotomized according to Hollingshead index score) as covari-
ates varying between subjects. We also examined the individual and unique
effects of negative, positive, and bored moods and perceived negativity of
interpersonal interaction on BP.
After the analyses determining the covariates of ambulatory BP, we
carried out random regression analyses determining the individual and
combined effects of dispositional optimism and trait anxiety on BP, along
with their interactions with negative mood. In addition to conducting
separate analyses for SBP and DBP, we conducted separate analyses to
determine the individual and combined effects of overall optimism, opti-
mism and pessimism subscales, and trait anxiety on cardiovascular activity.
With the within-subject, time-varying covariates, gender, and level of
occupational prestige in the model, personality measure, negative mood,
and their interactions were entered. The personality measure, dichotomized
with men and women combined, served as a between-subjects variable in
the model, whereas the ratings of negative mood varied within subjects.
Separate slopes and intercepts were computed for individuals scoring
above and below the median on overall optimism, on the optimism and
pessimism subscales, and on trait anxiety, and contrast comparisons were
computed to examine whether the within-subject effects of negative mood
on cardiovascular activity varied between the two groups.
The second major goal of our study was to test the individual and unique
effects of dispositional optimism and anxiety on ongoing states of mood.
Negative, positive, and bored moods and type of interpersonal interaction
(0 = neutral or positive, 1 = negative interaction or an argument) were
analyzed in separate models, as were the effects of overall optimism, the
optimism and pessimism subscales, and trait anxiety.
We used the maximum-likelihood estimation method and set up the
covariance matrix corresponding to the random effects in the model (i.e.,
RAIKKONEN, MATTHEWS, FLORY, OWENS, AND GUMP
variance components), assuming complete independence across the partic-
ipants. The within-subject residual covariance matrix was modeled accord-
ing to the first-order autoregressive errors. That is, the participant's re-
sponse at a given time point was assumed to be predicted by his or her
response at the adjacent time point and to a much lesser degree by a more
distal response. In each model, the average within-subject serial correlation
of residuals was substantial and significant (ps < .001 for all correlations).
Table 1 presents the mean values for the primary variables.
None of the dispositions varied by gender or level of occupational
prestige (ps > .10). The Pearson correlation coefficients were .53
between overall optimism score and trait anxiety and —.50 and .44,
respectively, between the optimism and pessimism subscales and
trait anxiety (ps < .001). The subscale scores were also signifi-
cantly related to one another (r = —.47, p < .001).
As reported elsewhere (Raikkonen, Matthews, Flory, & Owens,
1999), the within-subject correlations (Michela, 1990) between
mood and negative interpersonal interactions showed that the more
negative the participant's mood, the less positive she or he felt (r =
— .36, p < .001) and the greater the concurrent negativity of
interpersonal interactions (r = .56, p < .001). The more positive
the participant felt, the less she or he felt bored (r = — .32, p < .01)
and the less negative the concurrent interpersonal interaction (r =
—.30, p < .01). Correlations of bored mood with negative mood
(r = .03) and negative interpersonal interactions (r = .04) were
Covariates of Ambulatory Blood Pressure
The results of the random regression models testing the effects
of posture, location, physical activity, substance use, and interper-
sonal interaction are reported in Table 2. Analyses showed that
posture, location, and physical activity were consistently signifi-
cantly associated with SBP and DBP, with SBP being higher when
the participant was engaging in moderate or strenuous exercise,
driving a car, and sitting or standing. DBP was higher when the
participant was sitting or standing. Use of alcohol had an effect on
SBP only, whereas use of over-the-counter medications, caffeine,
and interpersonal interaction had no effects. Furthermore, men had
higher average SBP and DBP than women, but level of occupa-
tional prestige had no effects on BP. The subsequent random
regression models were conducted with location, posture, and
physical activity as covariates varying within subjects and gender
and occupational prestige as covariates varying between subjects.
The analyses testing the effects of moods and negative interper-
sonal interaction on BP are reported in Table 2. Negative mood
was associated with elevations in SBP, positive mood was asso-
ciated with elevations in SBP and DBP, and boredom was asso-
ciated with attenuated SBP and DBP. Negative interpersonal in-
teraction had no cardiovascular effects.
Influence of Dispositional Optimism, Trait Anxiety, and
Mood on Ambulatory Blood Pressure
Individual effects of traits. The effects of traits taken individ-
ually are presented in Table 3. Pessimists (according to overall
LOT score) had significantly higher average SBP and DBP levels
Predictors of Ambulatory Blood Pressure
Physical activity (vs. low)
Location (vs. home)
Posture (vs. reclining)
Alcohol (yes vs. no)
Caffeine (yes vs. no)
(yes vs. no)
Interpersonal social interaction
(yes vs. no)
Women (0) vs. men (l)a
Low (1) vs. high (0) job
a Estimate reflects a difference between the groups.
with physical activity, location, posture, gender, and occupational prestige
in the "model.
b Effect estimated
than optimists. This pattern was replicated with the optimism
subscale but not with the pessimism subscale.
The analyses testing the effects of mood on ambulatory BP indi-
cated that pessimists (as indicated by overall LOT score) showed
equivalent BP, regardless of the presence or absence of negative
mood: SBP, estimate = .20, r(5267) = 1.53, p = .13, and DBP,
estimate = -.10, f(5267) = -1.07, p = .29. In contrast, only when
in a negative mood did optimists (as indicated by overall LOT score)
show elevations in SBP, estimate = .44, z(5267) = 2.80, p = .005,
and DBP, estimate = .32, r(5267) = 2.83, p = .005 (see Figures 1 and
2). Contrast comparisons between optimists and pessimists assessed
by the overall LOT were not significant in the analysis of SBP, F(l,
5267) = 1.46, p = .23, but were significant in the analysis of DBP,
F(l, 5267) = 8.33,/? = .004. An identical pattern of significant results
was obtained in the analyses classifying participants via the pessi-
mism and optimism subscales.
Table 3 shows that individuals with high scores on the Spiel-
berger Trait Anxiety Inventory had significantly higher average
SBP and DBP levels than individuals with low scores. The BP
levels of those with high trait anxiety scores were unaffected by
negative mood: SBP, estimate = .07, r(5267) = .52, p = .60, and
DBP, estimate = -.16, f(5267) = -UQ,p
BP levels of those with low trait anxiety scores were elevated at
times of negative mood: SBP, estimate = .58, r(5267) = 3.87, p <
.001, and DBP, estimate = .36, t = 3.32, p < .001. Contrast
comparisons of the slopes between individuals with low and high
anxiety scores were significant in the analyses of both SBP, F(l,
< .09. Conversely, the
OPTIMISM AND AMBULATORY BLOOD PRESSURE
Average Ambulatory Blood Pressure Across 3 Days in Optimists, Pessimists,
and Anxious Participants
Systolic blood pressure
' Adjusted for anxiety-optimism.
5267) = 6.63, p < .01, and DBP, F(l, 5267) = 13.17, p < .0003
(see Figures 3 and 4).
Because dispositional optimism and trait anxiety may interact
with other moods or situational factors in determining BP, we
tested interactions of the dispositions with positive and bored
mood, interpersonal interaction, negative interpersonal interaction
(positive-neutral vs. negative-conflict), and home versus work
location. The results of these analyses showed no significant
interactions for overall LOT score, LOT subscale scores, or Spiel-
berger Trait Anxiety Inventory scores (p > .07).
Unique effects of traits. The analyses examining the indepen-
dent effects of optimism and trait anxiety on BP demonstrated that
the main effects of the overall LOT and the optimism subscale
scores were nonsignificant in the analyses of SBP after adjustment
for trait anxiety (ps > .12). The effects on DBP remained signif-
icant (ps < .02) and showed that pessimists (as indicated by
overall LOT score) and low optimists (according to the optimism
subscale) had higher DBP levels than their counterparts. Statistical
controls for overall LOT score or for optimism and pessimism
subscale scores (data not shown) did not alter the associations
between trait anxiety and BP (ps < .01).
Influence of Dispositional Optimism and Trait Anxiety on
Mood and Perceived Negativity of Interactions
Individual effects of traits. Table 4 shows that pessimists (ac-
cording to overall LOT score) reported higher negative mood,
lower positive mood, and more frequent negative interpersonal
interactions than did optimists (ps < .03). Individuals with high
scores on the pessimism subscale felt more negatively and reported
more frequent negative interpersonal interactions than individuals
with low pessimism scores (ps < .002). Individuals with high
optimism subscale scores felt less negatively, experienced fewer
negative interactions, and felt more positively than individuals
3 122 -
< - Optimists
35 72 i
Figure 1. Association between pessimism and diary ratings of negative
mood on ambulatory systolic blood pressure.
Figure 2. Association between pessimism and diary ratings of negative
mood on ambulatory diastolic blood pressure.
RAIKKONEN, MATTHEWS, FLORY, OWENS, AND GUMP
- • - Low Anxiety
—*— High Anxiety
Figure 3. Association between anxiety and diary ratings of negative
mood on ambulatory systolic blood pressure.
with low optimism scores (ps < .052). Table 4 shows that indi-
viduals scoring high on the Spielberger Trait Anxiety Inventory
experienced more negative and less positive mood and felt more
bored, tired, and apathetic than their low anxiety counterparts
across the days of monitoring (ps < .02).
We also examined whether the effects of optimism-pessimism
and anxiety on moods would be moderated by the occurrence of
interpersonal interactions in general, the occurrence of negative
interpersonal interactions, or home versus work location. The
benefits of interpersonal interactions for positive mood were less
pronounced for pessimists (according to overall LOT score), esti-
mate = 0.48, f(6980) = 4.08, p < .001, and for anxious individ-
uals, estimate = 0.48, t(69SQ) = 3.98, p < .001, than for their
more optimistic, estimate = 0.83, f(6980) = 7.51, p < .001, and
less anxious, estimate = 0.83, f(6980) = 7.44, p < .001, counter-
parts: overall LOT, F(l, 6980) = 4.41, p < .04, and anxiety, F(l,
6980) = 4.48, p < .03. There were no other significant
Unique effects of traits. Table 4 presents the results of the
analyses of the independent influences of traits. After control for
Spielberger Trait Anxiety Inventory scores, pessimists on the
overall LOT as well as on the pessimism subscale felt more
negatively and experienced more negative interpersonal interac-
tions than their more optimistic counterparts, whereas those with
high scores on the optimism subscale felt more positively than
individuals with low scores. The effect of overall LOT score was
nonsignificant in the analyses of positive mood, as were the effects
of the optimism subscale in the analyses of negative mood and
negative interpersonal interactions when statistical adjustment was
made for trait anxiety. When overall LOT scores were controlled,
higher trait anxiety scores were associated with more negative and
Other relevant data. We analyzed whether dispositional opti-
mism and trait anxiety would affect mood ratings measured at the
end of each of the 3 days. Pessimists on the overall LOT rated their
daily mood as more negative, t = 2.68, p = .01, and less positive,
t = — 2.25, p = .03, than optimists. With regard to the subscales,
pessimists rated their daily mood as more negative than their
counterparts, and optimists rated their daily mood as less negative
and more positive. Dispositional optimism had no effects on bore-
dom. Pessimists, determined by higher scores on either the overall
scale or the pessimism subscale, perceived their most stressful or
difficult events across the 3 days as more severe than their more
optimistic counterparts, ?s(98) > 2.12, ps < .04. Individuals high
in trait anxiety rated their daily mood as more negative, f(98) =
2.35, p < .02; less positive, r(98) = 3.25, p < .002; and more
bored, /(98) = 1.80, p < .08. Trait anxiety had no effects on
perceived severity of the most stressful event across the 3 days.
Dispositional optimism and trait anxiety were not related to
BP levels measured in the initial laboratory assessment session
(ps > .20).
This study sought to determine whether dispositional measures
of optimism, pessimism, and anxiety would affect ambulatory BP
and mood monitored over a wide range of circumstances during
routine everyday activities and whether any associations between
these dispositions and BP would be moderated by mood. The
findings indicated that pessimists routinely felt more negative, and,
when in an interpersonal interaction, they experienced it as more
negative, whereas optimists felt more positive. Pessimists, espe-
cially those with low optimism subscale scores, had higher overall
SBP and DBP levels than their counterparts throughout the 3 days,
even when they were not in a negative mood. However, when
optimists and low pessimists did experience negative moods (al-
beit infrequently), they exhibited BP as high as that observed in
pessimists and in low optimists. This may have been due to their
well-documented efforts to use problem-focused strategies of cop-
ing with stress (for a review, see Scheier & Carver, 1985, 1987,
1992), which may cause BP and heart rate to be elevated during
acute challenges (e.g., Sherwood, Dolan, & Light, 1990). Because
of our analytic strategy, our results cannot be attributed to variation
in posture, location, or physical activity of the optimists and
pessimists at the time of the measurement or to gender or occu-
pational prestige. Thus, the results confirm the study hypotheses
and add to the existing literature by showing that pessimism
confers adverse effects on psychological as well as physical well-
- • - Low Anxiety
—*— High Anxiety
Figure 4. Association between anxiety and diary ratings of negative
mood on ambulatory diastolic blood pressure.
OPTIMISM AND AMBULATORY BLOOD PRESSURE
Average Ratings of Mood Across 3 Days in Optimists, Pessimists, and Anxious Participants
Negative moodPositive mood Bored mood
p M p M p M p M
"Adjusted for anxiety-optimism.
being, not only during highly stressful circumstances but across a
variety of circumstances during routine everyday events.
The results from the analyses of trait anxiety are also potentially
important. Similar to the results on optimism-pessimism, our data
demonstrated that anxious individuals exhibited higher average
levels of BP and reported feeling more negatively, less positively,
and more bored during daily living than those with low anxiety
levels. The BP levels of anxious individuals were elevated regard-
less of the presence or absence of negative mood, whereas low
anxious individuals showed elevations in BP only when they were
in a negative mood. Again, these associations were independent of
fluctuations in posture, location, and physical activity, as well as
independent of gender and occupational prestige. Although it has
been argued that negative affectivity or neuroticism affects sub-
jective health complaints and illness behavior but may not affect
long-term health status (Watson & Pennebaker, 1989), our data do
suggest that neuroticism, at least as measured by the Spielberger
Trait Anxiety Inventory, is related to elevated BP throughout the
day (even though it was not shown to be related to resting pressure
measures in the laboratory). Perhaps the persistent elevations of
BP and emotional responding of anxious individuals lead to future
hypertension (Markovitz et al., 1991, 1993). If so, our data would
suggest that neuroticism does, in fact, affect health directly.
From a psychophysiological perspective, giving up and turning
away, characteristic of pessimists, and continued striving, character-
istic of optimists (Scheier & Carver, 1985, 1987, 1992), are the
defining features of defeat-type and defense-type reactions to stress,
respectively (see Cannon, 1929; Henry & Stephens, 1977; Selye,
1946). Similarly, individuals high in trait anxiety attempt to alter the
source of stress by passively avoiding or withdrawing from the
situation rather than by attempting to actively cope with the situation
(Smith et al., 1989). The defeat and defense reactions are two ex-
tremes of biological reactions to stress with different pathophysiolog-
ical and metabolic profiles. It has been hypothesized that although the
defense type of reaction to stress has effects, the physiological and
metabolic influences of a defeat reaction are most consequential (for
further discussion, see Bjorntorp, 1990). It has even been speculated
that the more passive stance in defeat reaction may play a role in the
development of primary hypertension (Henry & Grim, 1990). This
perspective should be taken not to suggest that optimism-pessimism
and anxiety be conceptualized by the two extremes of reaction pat-
terns to stress but to offer further ideas about the physiological
mechanisms by which these dispositions might confer adverse effects
Regarding the discriminant validity of optimism and pessimism,
our study findings demonstrated that optimism and pessimism
contributed significantly and independently of trait anxiety to
negative mood, perceived negative interactions, and diastolic BP
alterations during everyday living. Anxiety contributed signifi-
cantly and independently of optimism-pessimism to negative and
bored moods and BP alterations during everyday living. Taken
together, these findings suggest that dispositional optimism and
anxiety have unique, albeit relatively similar, effects on psycho-
logical and physical well-being. The independent effects of opti-
mism and anxiety obtained in the present study have been reported
in previous literature (e.g., Bromberger & Matthews, 1996; Mar-
shall & Lang, 1990; Robbins et al., 1991; Scheier et al., 1994).
The present study also evaluated whether the effects of dispo-
sitional optimism could be attributed to scores on the LOT pessi-
mism or optimism subscales or their sum (total score). It appeared
that the BP elevations of pessimists on the overall LOT were most
apparent among those low on the optimism subscale. Negative
mood states were elevated among both those high in pessimism
and those low in optimism, but statistical adjustments for trait
anxiety reduced the associations with optimism to nonsignificance.
Positive mood states were elevated among those high in optimism
but not those low in pessimism. These findings suggest that it
would be fruitful to continue to examine associations with the total
and subscale scores in future work so as to allow assessment of the
similarity and dissimilarity of effects of high positive and low
negative expectancies. Stated differently, the psychological and
physiological consequences of endorsing low positive and high
negative expectancies may be different.
There are a few notable limitations to our study. First, assessments
RAIKKONEN, MATTHEWS, FLORY, OWENS, AND GUMP
were made at regular time intervals (every 30 min) rather than
randomly. Random time intervals between assessments would prob-
ably reduce any anticipatory effect of the upcoming assessment and
would, therefore, be less likely to cause individuals to deviate from
their usual activities (Stone & Shiftman, 1994). Wearing of the
ambulatory monitor may have had some unique effects on the results
(Blanchard, Cornish, Wittrock, & Jaccard, 1990), such as participants
avoiding situations that would cause embarrassment because of the
appearance of the monitor. However, unlike the study by Blanchard et
al. (1990), our study used a silent ambulatory monitoring device, the
measurement being less objectionable to individuals when they are
wearing the monitor in the company of others and, therefore, less
likely to cause them to depart from their normal behavioral patterns
(Matthews et al., 1992). Despite this possibility, there is no reason to
presume that any effects of anticipation or of wearing the monitor
vary between optimists and pessimists or between those high and low
in terms of anxiousness. Therefore, we find it unlikely that wearing
the monitor and anticipating the measurement would bias the results
in any systematic manner. Second, we examined normotensive
middle-aged men and women without any chronic diseases and who
did not use medications with cardiovascular and metabolic effects.
The homogeneity of the health of the sample may, however, be more
of a benefit than a limitation. At least, it is likely that the associations
were not inflated because of the somewhat restricted variation in
ambulatory BP levels.
Despite these possible limitations, there are a number of
strengths of the study design and data-analytic strategy applied to
address the research questions. An equal number of employed
middle-aged men and women were recruited, and men and women
were matched for their level of occupational prestige. A relatively
large number of observations were made with relatively few miss-
ing values, strengthening the generalizability of the findings. The
issue of generalizability extends to the present data-analytic strat-
egy in which an individual was treated as a random variable (see
Schwartz et al., 1994). That is, we did not assume that the within-
subject effects of position, location, physical activity, or mood on
BP and heart rate were identical for everyone or confounded by
between-subjects effects. In such a case, investigators can legiti-
mately make inferences about the larger population from which
their samples are selected (Schwartz et al., 1994). The multilevel
random regression analysis enabled the data to be evaluated at both
the idiographic and the nomothetic level. At the idiographic level
we examined intraindividual variability in mood and BP, and at the
nomothetic level we examined interindividual differences in intra-
individual variability between optimists and pessimists and be-
tween individuals high and low in terms of anxiousness. Thus,
among the particular advantages of the multilevel random regres-
sion analysis approach used were the assumption of interperson
variability in slopes and intercepts and the possibility of acknowl-
edging the problem of correlated errors in a repeated measures
design in which multiple observations are derived from the same
participant over time.
In summary, using a sophisticated analytic technique, we found
that optimists and pessimists had different experiences during the 3
days of monitoring, as did those who were high and low in terms of
anxiousness. Presumably because of their general expectations about
the likelihood of positive and negative outcomes, pessimists experi-
enced high negative and low positive moods and had high ambulatory
BP levels during their daily activities. Optimists experienced high
positive and low negative moods and had low ambulatory BP levels
except on the infrequent occasions when they were experiencing a
negative mood. Anxious individuals experienced high negative
moods, were bored, and had elevated BP levels. These experiences,
should they occur repeatedly in their lives, would make optimists
likely to live long, healthy lives and pessimists and those who are
anxious likely to live short, unhealthy ones.
Bjorntorp, P. (1990). Visceral fat accumulation: The missing link between
psychosocial factors and cardiovascular disease. Journal of Internal
Medicine, 230, 195-201.
Blanchard, E. B., Cornish, P. J., Wittrock, D. A., & Jaccard, J. (1990).
Effects of 24-hour ambulatory blood pressure monitoring on daily ac-
tivities. Health Psychology, 9, 647-652.
Bolger, N. (1990). Coping as a personality process: A prospective study.
Journal of Personality and Social Psychology, 59, 525-537.
Bolger, N., & Schilling, E. A. (1991). Personality and the problems of
everyday life: The role of neuroticism in exposure and reactivity to daily
stressors. Journal of Personality, 59, 355-386.
Bromberger, J. T., & Matthews, K. A. (1996). A longitudinal study of the
effects of pessimism, trait anxiety, and life stress on depressive symp-
toms in middle-aged women. Psychology and Aging, 11, 207-213.
Cannon, W. R. (1929). Bodily changes in pain, hunger, fear and rage (2nd
ed.). New York: Appleton.
Carver, C. S., & Gaines, J. C. (1987). Optimism, pessimism, and postpar-
tum depression. Cognitive Therapy and Research, 11, 449-462.
Carver, C. S., Pozo, C, Harris, S. D., Noriega, V., Scheier, M. F.,
Robinson, D. S., Ketcham, A. S., Moffat, F. J., Jr., & Clark, K. C.
(1993). How coping mediates the effect of optimism on distress: A study
of wbinen with early stage breast cancer. Journal of Personality and
Social Psychology, 65, 375-390.
Devereux, R. B., Pickering, T. G., Harshfield, G. A., Kleinert, H. D.,
Denby, L., Clark, L., Pergibon, D., Jason, M., Kleiner, R., Borer, J. S.,
& Laragh, J. H. (1983). Left ventricular hypertrophy in patients with
hypertension: Importance of blood pressure response to regularly recur-
ring stress. Circulation, 68, 470-476.
Eysenck, S. B. G., & Eysenck, H. J. (1968). The measurement of psychoti-
cism: A study of factor stability and reliability. British Journal of Social
and Clinical Psychology, 7, 286-294.
Fitzgerald, T. E., Tennen, H., Affleck, G., & Pransky, G. S. (1993). The
relative importance of dispositional optimism and control appraisals in
quality of life after coronary artery bypass surgery. Journal of Behav-
ioral Medicine, 16, 25-43.
Hedges, S. M., Krantz, D. S., Contrada, R. C, & Rozanski, A. (1990).
Development of a diary for use with ambulatory monitoring of mood,
activity, and physiological function. Journal of Psychopathology and
Behavioral Assessment, 12, 203-217.
Henry, J. P., & Grim, C. E. (1990). Psychosocial mechanisms of primary
hypertension. Current Science, 8, 783-793.
Henry, J. P., & Stephens, P. M. (1977). Stress, health and the social
environment. A sociobiologic approach to medicine. New York:
Larsen, R. J., & Kasimatis, M. (1991). Day-to-day physical symptoms:
Individual differences in the occurrence, duration, and emotional con-
comitants of minor daily illnesses. Journal of Personality, 59, 387-423.
Larsen, R. J., & Ketelaar, T. (1991). Personality and susceptibility to
positive and negative emotional states. Journal of Personality and Social
Psychology, 61, 132-140.
Markovitz, J. H., Matthews, K. A., Kannel, W. B., Cobb, J. L., &
D'Agostino, R. B. (1993). Psychological predictors of hypertension in
the Framingham Study. Is there tension in hypertension? Journal of the
American Medical Association, 270, 2439-2443.
OPTIMISM AND AMBULATORY BLOOD PRESSURE Download full-text
Markovitz, J. H., Matthews, K. A., Wing, R. R., Kuller, L. H., & Meilahn,
E. N. (1991). Psychological, biological, and health behavior predictors
of blood pressure changes in middle-aged women. Journal of Hyperten-
sion, 9, 399-406.
Marshall, G. N., & Lang, E. L. (1990). Optimism, self-mastery, and
symptoms of depression in women professionals. Journal of Personality
and Social Psychology, 59, 132-139.
Marshall, G. N., Wortman, C. B., Kusulas, J. W., Hervig, L. K., & Vickers,
R. R., Jr. (1992). Distinguishing optimism from pessimism: Relations to
fundamental dimensions of mood and personality. Journal of Personal-
ity and Social Psychology, 62, 1067-1074.
Matthews, K. A., Owens, J. F., Allen, M. T., & Stoney, C. M. (1992). Do
cardiovascular responses to laboratory stress relate to ambulatory blood
pressure levels?: Yes, in some of the people, some of the time. Psycho-
somatic Medicine, 54, 686-697.
Michela, J. L. (1990). Within-person correlational design and analysis. In
C. Hendrick & M. S. Clark (Eds.), Research methods in personality and
social psychology (pp. 279—311). Newbury Park, CA: Sage.
Mroczek, D. K., Spiro, A., m, Aldwin, C. M., Ozer, D. J., & Bosse, R. (1993).
Construct validation of optimism and pessimism in older men: Findings
from the Normative Aging Study. Health Psychology, 12, 406-409.
Perloff, D., Sokolow, M., & Cowan, R. (1983). The prognostic value of
ambulatory blood pressure. Journal of the American Medical Associa-
tion, 249, 2792-2798.
PROC MIXED. SAS/STAT software: Changes and enhancements through
release 6.12 [Computer software]. (1997). Cary, NC: SAS Institute.
Raikkonen, K., Matthews, K. A., Flory, J. D., & Owens, J. F. (1999). The
effects of hostility on ambulatory blood pressure and mood during daily
living in healthy adults. Health Psychology, 18, 44-62.
Robbins, A. S., Spence, J. T., & Clark, H. (1991). Psychological determi-
nants of health and performance: The tangled web of desirable and
undesirable characteristics. Journal of Personality and Social Psychol-
ogy, 61, 755-765.
Robinson-Whelen, S., Kim, C, MacCallum, R. C, & Kiecolt-Glaser, J. K.
(1997). Distinguishing optimism from pessimism in older adults: Is it
more important to be optimistic or not to be pessimistic? Journal of
Personality and Social Psychology, 71, 1345-1353.
Scheier, M. F., & Carver, C. S. (1985). Optimism, coping, and health:
Assessment and implications of generalized outcome expectancies.
Health Psychology, 4, 219-247.
Scheier, M. F., & Carver, C. S. (1987). Dispositional optimism and
physical well-being: The influence of generalized outcome expectancies
on health. Journal of Personality, 55, 169-210.
Scheier, M. F., & Carver, C. S. (1992). Effects of optimism on psycho-
logical and physical well-being: Theoretical overview and empirical
update. Cognitive Therapy and Research, 16, 201-228.
Scheier, M. F., Carver, C. S., & Bridges, M. W. .(1994). Distinguishing
optimism from neuroticism (and trait anxiety, self-mastery, and self-
esteem): A reevaluation of the Life Orientation Test. Journal of Person-
ality and Social Psychology, 67, 1063-1078.
Scheier, M. F., Matthews, K. A., Owens, J. F., Magovern, G. J., Sr.,
Lefebvre, R. C, Abbott, R. A., & Carver, C. S. (1989). Dispositional
optimism and recovery from coronary artery surgery: The beneficial
effects on physical and psychological well-being. Journal of Personality
and Social Psychology, 57, 1024-1040.
Schulz, R., Bookwala, J., Knapp, J. E., Scheier, M., & Williamson, G. M.
(1996). Pessimism, age, and cancer mortality. Psychology and Ag-
ing, 11, 304-309.
Schwartz, J. E., Warren, K., & Pickering, T. G. (1994). Mood, location and
physical position as predictors of ambulatory blood pressure and heart
rate: Application of a multi-level random effects model. Annals of
Behavioral Medicine, 16, 210-220.
Selye, H. (1946). The general adaptation syndrome and the diseases of
adaptation. Journal of Clinical Endocrinology, 6, 217-230.
Sherwood, A., Dolan, C. A., & Light, K. C- (1990). Hemodynamics of
blood pressure responses during active and passive coping. Psychophys-
iology, 27, 656-668.
Smith, T. W., Pope, K. M., Rhodewalt, F., & Poulton, J. L. (1989).
Optimism, neuroticism, coping, and symptom reports: An alternative
interpretation of the Life Orientation Test. Journal of Personality and
Social Psychology, 56, 640-648.
Sokolow, M., Werdegar, D., Kain, H. K., & Hinman, A. T. (1966).
Relationship between level of blood pressure measured casually and by
portable recorders and severity of complications in essential hyperten-
sion. Circulation, 34, 279-298.
Spielberger, C. D. (1983). Manual for the State-Trait Anxiety Inventory.
Palo Alto, CA: Consulting Psychologists Press.
Stone, A. A., & Shiffman, S. (1994). Ecological momentary assessment
(EMA)-in behavioral medicine. Annals of Behavioral Medicine, 16,
Watson, D., & Clark, A. L. (1984). Negative affectivity: The disposition to
experience aversive emotional states. Psychological Bulletin, 96, 465-490.
Watson, D., & Pennebaker, J. W. (1989). Health complaints, stress, and
distress: Exploring the central role of negative affectivity. Psychological
Review, 96, 235-254.
Williams, R. D., Riels, A. G., & Roper, K. A. (1990). Optimism and
distractibility in cardiovascular reactivity. Psychological Record, 40,
Received April 21, 1997
Revision received February 2, 1998
Accepted May 25, 1998